Global Sources
EE Times-Asia
Stay in touch with EE Times Asia
EE Times-Asia > Interface

Signature authentication brings new opportunities

Posted: 01 Apr 2003 ?? ?Print Version ?Bookmark and Share

Keywords:biometric? biometric signature authentication system? identification technology? security system? ibg?

Biometric signature authentication can identify a person just by analyzing the signature. This can be done by using an electronic pen and extracting characteristic vectors of the signature. Each person has a different handwriting, which can be recognized. During signing, dynamic information like velocity, time, stroke order and writing pressure occur as well as static information like shape.

Signature is a representation of features carved in the human brain by long-continued habitual behavior. It has consistence within certain range and does not change in a short period. In contrast to other identification methods that only rely on one identifier, biometric signature authentication performs more than five identification steps at once. Even though static information can be imitated, this kind of information is hardly copied because it is not visible, and it is too difficult to imitate more than five behavioral characteristics.

However, there are concerns on the accuracy of this technology since no one can make an exact signature every time. The trouble comes when the system rejects a true signature (false rejection or FR) or accepts a false signature (false acceptance or FA). To increase accuracy, we have to reduce both cases. The lower the cross point (FR=FA point) is, the better the system becomes. None of the two can be neglected. Instead, it should be compromised depending on the application. For instance, when a higher security level is needed, setting the threshold lower reduces FA error in the cost of FR. Consequently, when a relatively lower security level is needed, setting the threshold higher reduces FR rate.

There are three approaches in comparing target signature with references. First is global feature comparison, which can generate statistic model easily. However, this does not show local difference well. Second is point matching, which can compare local details but does not work well in utilizing the structural information of the signature. The third approach is article comparison. This is a technique to compare the article to its counterpart after separating. It uses a statistical model of each article.

Signature authentication benefits

A signature is a unique behavioral characteristic of a person that cannot be given away or performed by other people involuntarily. Unlike passwords, it cannot be forgotten or given to others, nor does it give emotional rejection and invade personal privacy. Even when hacked, it does not disable the system for a long time.

Biometric signature authentication system can easily be used in conventional identification media such as credit card, banking, immigration control and electronic document. And unlike other biometric systems that need signature affixing after identification, this system generates the effect of signature at a spur.

One of the obstacles that hindered the growth of the biometrics market is the need for additional input devices and the high price of the equipment. Fortunately, devices these days are more affordable because of the increase in production and the advances in technology. Small tablet and pen digitizer, the input device of the signature authentication system, are highly cost competitive as compared to other input devices. Besides, the advent of PDA and IMT-2000 removes the need for a tablet.

Promising niche market

The deployment of the signature authentication system is rather low at the moment. However, according to the International Biometric Group (IBG) report, the revenue of global biometric signature authentication is expected to grow from $3 million in 2000 to $111 million in 2005. This corresponds to 5 percent of overall biometrics market.

Among foreign companies, Communication Intelligence Corp., Cyber-SIGN, Sign-On Systems, SoftPro, Analog, and LCI Group have already rolled out their signature authentication engines. In Korean market, Passsign, Mmi Group, and Diotec also rolled out their own signature authentication engines: PS2000 & PS2001CE, SignQ, and DioPass.

- Joh Yoon-Ju

Electronic Engineering Times - Asia

Article Comments - Signature authentication brings new ...
*? You can enter [0] more charecters.
*Verify code:


Visit Asia Webinars to learn about the latest in technology and get practical design tips.

Back to Top